Re simulating splash and foam from and existing liquid sim

[jeff_grigsby]

I have an existing fluid wave simulation with no splash or foam,, and I ran a re simulation to add in these elements yielding 40 million particles. If I were to do this 2x, with the exact same settings, would Phoenix produce the exact same result, or would they be slightly different? The reason I ask is I would like to increase the splash, mist, and foam simulated particle count from 40 million,, to 200 million by splitting the process I mentioned above across the network 5x and re assembling them in a .PRT format, assuming the results vary slightly each time you run a re simulation..

if not, Add to the wish list, a random seed variable to split particle generation in parallel across numerous machines.

Thanks
Jeff

[Svetlin.Nikolov]

Hey,

Indeed, it will be slightly different each time, which is a big pain for me, but surprisingly in this case it's a positive thing :O

Sounds like you should be able to assemble the particles from several simulations without them overlapping.

Cheers!

[jeff_grigsby]

Thanks Svetin,

this is good news for sure. It seems this is a good path for multiple CPU parallel processing of particles, now only if I could get the initial ocean sim split across several machines.

[Svetlin.Nikolov]

Ahhh yes, this will be a lot harder, but still a very clever approach! Would be great if we automate this...